megaloblastosis: from morphos to...

Fax +41 61 306 12 34E-Mail [email protected]

Review

Med Princ Pract 2005;14(suppl 1):2–14 DOI: 10.1159/000086179

Megaloblastosis: From Morphos to Molecules

K.C. Das

a Monisha Das

b D. Mohanty

b M.M. Jadaon

a A. Gupta

a R. Marouf

a S.K. Easow

a

a Hematology Unit, Department of Pathology, Faculty of Medicine, Kuwait University, Kuwait;

b Institute of Immunohematology, Mumbai , India

thesis occurred due to an impaired de novo pathway of thymidylate synthesis in both vitamin-B 12 - and folate-de-fi cient human megaloblastic bone marrows as well as in the bone marrows of rhesus monkeys and rats with ex-perimentally induced folate defi ciency. Interestingly, fo-late-defi cient monkeys developed frank megaloblastic bone marrows, but folate-defi cient rats did not. On the other hand, megaloblastic changes in the bone marrow of human patients with myelodysplastic syndrome and erythroleukemia were not associated with this DNA syn-thetic abnormality. Biosynthesis of predominantly argi-nine-rich histones in megaloblastic bone marrows was markedly reduced as compared to normoblastic bone marrows, which was consistently associated with elon-gation and despiralization of chromosomes and fi nely stippled nuclear chromatin in megaloblasts. Conclusion: The impaired biosynthesis of predominantly arginine-rich nuclear histones appeared to be a common molecu-lar event (a denominator) underlying the development of megaloblastosis with or without abnormal DNA syn-thesis.

Copyright © 2005 S. Karger AG, Basel

Introduction

Megaloblastic anemia is essentially recognized as a morphological entity and is characterized by megaloblas-tic transformation of erythroid precursor cells in the bone

Key Words Megaloblastosis � Electron microscopy � Chromosomal despiralization � Autoradiography � 3 H-thymidine incorporation � DNA synthesis � Deoxyuridine suppression test � Histone biosynthesis

Abstract Objective: Megaloblastosis (i.e., megaloblastic transfor-mation of erythroid precursor cells in the bone marrow) is the cytomorphological hallmark of megaloblastic ane-mia resulting from vitamin B 12 and folate defi ciency. It is characterized by a fi nely stippled lacy pattern of nuclear chromatin, which is believed to be an expression of de-ranged cellular DNA synthesis. However, the molecular basis of these cytomorphological aberrations still re-mains obscure. The current presentation describes the results of our studies on some molecular events associ-ated with the development of megaloblastosis. Meth-

ods: Transmission electron microscopy was used to study megaloblasts as well as DNA fi bers extracted from megaloblastic and normoblastic bone marrows with and without treatment with proteinase K during the extrac-tion procedure; cellular DNA synthesis in bone marrow cultures was studied by incorporation of 3 H-thymidine and deoxyuridine suppression test, while histone bio-synthesis in bone marrow cells was studied by in vitro incorporation of 3 H-tryptophan, 3 H-lysine and 3 H-argi-nine into histones. Results: Derangement of DNA syn-

Received: August 15, 2004 Revised: October 25, 2004

Dr. K.C. DasHematology Unit, Department of Pathology, Faculty of Medicine, Kuwait UniversityPO Box 24923, 13110 Safat (Kuwait)Tel. +965 531 9476, Fax +965 533 8905E-Mail [email protected], [email protected]

© 2005 S. Karger AG, Basel1011–7571/05/0147–0002$22.00/0

Accessible online at:www.karger.com/mpp

Dow

nloa

ded

by:

149.

126.

78.4

9 -

11/1

9/20

15 1

0:11

:04

PM

http://dx.doi.org/10.1159%2F000086179

Megaloblastosis Med Princ Pract 2005;14(suppl 1):2–14 3

marrow of these patients. Megaloblastosis is the cytomor-phological diagnostic marker of megaloblastic anemia, and represents a unique morphological and functional aberration of erythropoiesis. Megaloblasts are character-ized by several cytomorphological features which are dis-tinct from those of normal erythroid precursor cells (i.e., normoblasts) at all stages of maturity. These include larg-er cell size, and a fi nely stippled, open lacy pattern of nuclear chromatin. The cytoplasmic features are less re-markable, with evident asynchrony between nuclear and cytoplasmic maturation. The original prototype of this anemia was Addisonian pernicious anemia, initially rec-ognized in the Caucasians as the only example of mega-loblastic anemia in the nineteenth century. It was later discovered that this anemia was caused by a defi ciency of vitamin B 12 due to a lack of its obligatory binder gastric intrinsic factor and consequent malabsorption of this vi-tamin [1] . Lucy Wills [2, 3] described for the fi rst time another type of megaloblastic anemia, the ‘pernicious anemia of pregnancy’ and ‘tropical macrocytic anemia’, which were later found to be due to a defi ciency of folate (folic acid). It is, however, not infrequent to fi nd strict vegetarians and vegans in the Indian subcontinent (either due to religious and social compulsions, food faddism or poverty) who develop vitamin B 12 defi ciency and mega-loblastic anemia even when gastric intrinsic factor is not defi cient [4, 5] . Common causes of megaloblastic ane-mias are defi ciency of vitamin B 12 and folate due to var-ious reasons such as poor dietary intake, malabsorption, increased requirement, utilization and excretion [6] . Cy-tomorphological changes very similar to megaloblastosis can also occur in erythroid precursor cells in the bone marrow of patients with myelodysplastic syndrome and erythroleukemia; these latter conditions are relatively rare, and the nature of these ‘megaloblastic’ changes re-mains obscure. Whereas vitamin-B 12 - and folate-defi -cient megaloblastic anemias are successfully treated by providing the defi cient vitamins (i.e., folic acid or vitamin B 12 ), ‘megaloblastic’ changes associated with myelodys-plastic syndrome and erythroleukemia are resistant to treatment with these vitamins. Megaloblastic changes caused by vitamin B 12 and folate defi ciency appear to be systemic processes not restricted to the cells of the ery-throid series, but affecting all replicating cells [1, 7–9] . The appearance of giant metamyelocytes and hyperseg-mented neutrophils in vitamin-B 12 - and folate-defi cient megaloblastic anemias are also believed to be cytomor-phological expressions of these vitamin defi ciencies [1, 9] .

When peripheral blood lymphocytes of patients with vitamin-B 12 - or folate-defi cient megaloblastic anemias are stimulated by a mitogen such as phytohemagglutinin in tissue cultures, they undergo ‘blastic’ transformation, but the cytomorphological features of these transformed lymphoblasts are different from similarly transformed lymphoblasts from normal subjects. The transformed lymphoblasts from megaloblastic anemia patients are generally larger, possess fi nely stippled nuclear chroma-tin, and also develop a DNA synthetic abnormality as revealed by an abnormal or impaired de novo pathway of thymidylate synthesis. They have been described as ‘megaloblastic lymphocytes’ [10, 11] . The cytomorpho-logical changes in megaloblastic erythroid precursor cells and other replicating cell systems are generally believed to be morphological expressions of a poorly defi ned de-rangement of cellular DNA synthesis [7, 10–12] . This hy-pothesis has not since been adequately explored or veri-fi ed in clinical or experimental conditions.

The current presentation is based on our studies on cytomorphological features and molecular events in hu-man megaloblastic bone marrows, and in the bone mar-rows of two mammalian species (i.e., rhesus monkeys and Wistar strain rats) rendered folate defi cient by dietary deprivation of folate. These studies included (a) cytomor-phological studies of human megaloblastic bone marrows with particular focus on nuclear chromatin; (b) cytoge-netic analysis for chromosomal pattern and replication; DNA content, DNA synthesis and deoxyuridine (dU) suppression test; (c) cytomorphological features, DNA synthesis and dU suppression test in the bone marrows of folate-defi cient monkeys and rats, and (d) histone bio-synthesis in megaloblastic and control (normoblastic) bone marrows.

Materials and Methods

Human Subjects This study was carried out on 50 patients with megaloblastic

anemia of moderate to severe degree ( table 1 ). Among them, 40 had vitamin B 12 defi ciency (25 had pernicious anemia with positive Schilling’s test, parietal cell antibodies and intrinsic factor antibod-ies, whereas 15 had nutritional defi ciency of vitamin B 12 , being vegetarian or vegans) and 10 had folate defi ciency as indicated by subnormal serum and red cell folate levels either due to poor folate intake or malabsorption. There were 15 patients with myelodys-plastic syndrome, who presented with persistent pancytopenia, and macrocytic red cells with a few hypolobulated and hypogranular neutrophils (pseudo-Pelger anomaly). Their bone marrows showed normocellularity or hypercellularity and varying degrees of dysplas-tic changes involving two or more cell lines; the erythroid precursor cells in their bone marrows showed signifi cant, but varying degrees

Dow

nloa

ded

by:

149.

126.

78.4

9 -

11/1

9/20

15 1

0:11

:04

PM

Das /Das /Mohanty /Jadaon /Gupta /Marouf /Easow

Med Princ Pract 2005;14(suppl 1):2–14 4

of megaloblastic changes. There were 5 patients with erythroleuke-mia (AML, M6), which were diagnosed by their peripheral blood count, bone marrow cytomorphology, cytochemical features and immunophenotyping characteristics; their bone marrows showed megaloblastic changes. Thirty subjects aged 10–40 years, who were hematologically normal, but had undergone bone marrow examina-tion for unrelated disorders to exclude metabolic disorders (e.g. Gaucher’s disease, Niemann-Pick’s disease), or for chromosomal studies to exclude chromosomal mosaic, were studied as controls. Bone marrow aspiration and trephine biopsies were performed af-ter obtaining informed consents. Twenty patients with moderate to severe iron defi ciency anemia were studied for histone biosyn-thesis in the bone marrow and compared with those with megalo-blastic anemia.

Experimental Folate Defi ciency Folate defi ciency was produced in two mammalian species:

(a) higher mammal: rhesus monkeys; (b) lower mammal: Wistar strain rats.

Folate Defi ciency in Rhesus Monkeys. Five adult healthy rhesus monkeys (Macaca mulatta) each weighing 5–6 kg were caged indi-vidually in a well-lighted room, acclimatized to the laboratory con-ditions for 6–8 weeks. Folate defi ciency was produced in these an-imals by feeding them a folate-defi cient diet through a Ryle’s tube

as described by us previously [13] . Another 5 monkeys were sepa-rately caged and were fed a folate-rich diet (containing 100 � g folic acid per 100 g of the diet) to serve as the control group. The devel-opment of folate defi ciency was monitored by measuring the levels of folate in the serum, red cells and fi nally in the liver. The monkeys fed a folate-poor diet showed consistent evidences of development of folate defi ciency in 16–20 weeks, whereas the animals fed a fo-late-rich diet (the control group) showed no defi ciency ( table 2 ). At the end of the experiments, the monkeys were anesthetized, and in addition to collecting blood samples, bone marrows were aspirated from the posterior iliac spine for morphological studies, short-term tissue cultures, DNA synthesis, dU suppression test and other bio-chemical analyses.

Folate Defi ciency in Rats. Folate defi ciency was produced in 10 weanling (4 weeks old) Wistar strain rats weighing about 50–60 g each. Each rat was kept in a separate cage. Folate defi ciency was produced by feeding them a folate-defi cient diet as described previ-ously [13, 14] . Another 10 rats of similar age, sex and weight were separately caged to serve as the control group. The control animals were pair-fed with an isocaloric, folate-rich diet containing 50 mg of folic acid per kilogram of the diet; the folate-defi cient diet con-tained no measurable folic acid content. Preliminary experiments showed that the rats fed on this defi cient diet developed folate de-fi ciency after 12–16 weeks. This was indicated by consistently low

Table 1. Basic hematological profi le of 50 patients with megaloblastic anemia and 30 control subjects

Mean cell volume, fl

Megaloblastic changesin the bone marrow

Type of vitamin defi ciency

vitamin B12 folate

PatientsSeverely anemic, 110–118 ++++ n = 10 n = 5Hb 20–70 g/l (n = 15) 35–80 ng/l serum folate (0.5–2.5 �g/l)

red cell folate (50–86 �g/l)

Moderately anemic,Hb 71–96 g/l (25) (n = 35)

104–115 +++ n = 3055–97 ng/l

n = 5serum folate (1.0–3.0 �g/l)red cell folate (75.0–105 �g/l)

Control subjectsHb 140–160 g/l (n = 30) 78–86 none; all showed normo-

blastic erythropoiesis290–850 ng/l serum folate (8–20 �g/l)

red cell folate (350–980 �g/l)

Table 2. Folate levels in serum, red cells and liver before and after feeding rats and rhesus monkeys on folate-poor diet

Experimental groups Serum folate, �g/l Red cell folate, �g/l Liver folate, �g/g

initial after 16 weeks initial after 16 weeks initial after 16 weeks

Control rats (n = 10) 16.385.4 15.886.5 552.4810.2 485.4815.5 4.581.6 4.882.1Rats on folate-poor diet (n = 10) 17.286.1 2.580.5 470.587.5 83.480.7 4.781.2 0.880.3Control monkeys (n = 5) 19.285.2 18.586.3 436.4810.1 455.488.5 6.482.2 6.081.9Monkeys on folate-poor diet (n = 5) 18.486.4 2.580.3 476.489.5 76.588.4 5.981.7 1.280.6

Dow

nloa

ded

by:

149.

126.

78.4

9 -

11/1

9/20

15 1

0:11

:04

PM


levels of folate in the serum, red cells and the liver in these animals, whereas the control animals showed no signifi cant decrease in fo-late levels ( table 2 ). At the end of the experiments, the rats were anesthetized by administering anesthetic ether and sacrifi ced. The bone marrows were taken from the femur and processed for mor-phology, short-term tissue culture, DNA synthesis and dU suppres-sion test and other biochemical analyses.

Hematological Studies Routine hematological studies were performed by standard

techniques [15] . Blood and bone marrow smears were stained by May-Grünwald-Giemsa stain for cell morphology.

Transmission Electron Microscopy of Human Bone Marrow Cells Briefl y, the following procedure was used. Aspirated bone mar-

row cells were collected in balanced salt solution, centrifuged to make a cell button, which was fi xed in 3.0% glutaraldehyde solution for 2–3 h, washed in Millonig’s buffer, pH 7.3, and then treated with 1.0% osmium tetroxide, washed again in Millonig’s buffer for 3 h, appropriately dehydrated in increasing concentrations of etha-nol and dried in absolute ethanol, then put in two changes of pro-pylene oxide for 10 min each, and fi nally blocks were made in pure epoxy resin, from which ultra-thin sections were fi nally made for transmission electron microscopy in a JEOL 1200 EX.

DNA Synthesis and dU Suppression Test A portion of the aspirated bone marrow was collected asepti-

cally in cold heparinized 0.06 M Tris-buffered Hank’s balanced salt solution, pH 7.4, and processed into a monolayered suspension for short-term culture; DNA synthesis was monitored by measuring incorporation of 3 H-thymidine ( 3 H-TdR) into DNA, and dU sup-pression test was done as described by us previously [16, 17] . The dU suppression values were calculated as follows:

3

3H-TdR incorporation into DNA with dU

Percentage due suppression

100H-TdR incorporation into DNA without dU

=

×

3

dU suppression correction by the additions of folates or

100H-TdR incorporation into DNA without dU

×

3H-TdR incorporation into DNA with dU 12 and folate and/or vitamin B

12vitamin B =

3 H-TdR incorporation into cellular DNA is a sensitive radio-nucleoside marker of DNA synthesis. The generation of deoxythy-midine monophosphate (dTMP) and then deoxy-thymidine tri-phosphate (dTMP ] dTTP) is of particular importance for this syn-thetic process due to relative specifi city of the thymine base in the composition of DNA ( fi g. 8 ). The dU suppression test essentially measures the effi cacy of the de novo pathway by studying the in-corporation of 3 H-TdR into DNA (via the salvage pathway) in the presence of a standardized excess of dU added to the short-term tissue culture of bone marrow cells. Normally, the deoxyuridylate (dUMP) suppresses the incorporation of 3 H-TdR into cellular DNA to 10% of control cultures (i.e., cultures without added dU). In defi ciency of folate and vitamin B 12 , the de novo pathway is relatively inactive as a result of which the added dU fails to sup-

press the incorporation of 3 H-TdR into DNA (i.e., 3 H-TdR incor-poration is higher than 10% of the control). The higher the dUsuppression value, the more inactive or ineffi cient is the de novo pathway. This abnormality can be corrected to 10% or less of dU-suppressed values by the addition of the defi cient vitamins (folate or vitamin B 12 ). While all folates including 5-methyl-tetrahydrofo-late (THF) completely correct the dU suppression abnormality due to folate defi ciency, a combination of vitamin B 12 and 5-methyl-THF is required to correct the abnormality in vitamin B 12 defi -ciency [6, 10, 15] .

The incorporation of 3 H-TdR into individual cells was studied by autoradiography as described by us previously [18] . DNA con-tent of individual cells in the bone marrow smears was measured after Feulgen staining in a Becton-Dickinson Cellular Imaging Sys-tem (CAS 200). Cytogenetic analysis was done as previously de-scribed by us [19] .

Extraction and isolation of DNA from bone marrow cells were done by the classical phenol-chloroform extraction method as de-scribed by Sambrook et al. [20] . Replicate samples from each bone marrow were subjected to treatment with proteinase K during the extraction procedure in order to remove the nuclear proteins such as histones from the DNA, whereas the other sample from each bone marrow was not treated with proteinase K. The DNA fi bers were precipitated and dehydrated in ethanol and air-dried. These fi bers were then fi xed in 3.0% glutaraldehyde, washed in Millonig’s buffer, dehydrated in ethanol and fi nally embedded in epoxy resin, and ultra-thin sections were made for transmission electron micros-copy.

Histone Content and Histone Biosynthesis in Bone Marrow Cells Histone content was measured and its biosynthesis studied by

methods previously described [21–24] . The incorporations of 3 H-tryptophan, 3 H-lysine and 3 H-arginine into bone marrow cells in vitro were used as radioactive tracers for biosynthesis of his-tones.

Serum and Red Cell Folate and Vitamin B 12 Serum and red cell folate and vitamin B 12 were estimated by

radioisotope dilution techniques using appropriate binders for these two vitamins as described previously [25, 26] .

Statistical Method Statistical package for Social Sciences (SPSS Inc., Chicago, Ill.,

USA) was used for data processing; a p value ^0.05 was used as cutoff level of signifi cance. Student’s t test was used for comparing means.

Results

Cytomorphological Changes in Blood and Bone Marrows The patients with megaloblastic anemia due to defi -

ciency of vitamin B 12 or folate had variable degrees of anemia from moderate to severe, with or without throm-bocytopenia or leukopenia ( table 1 ). Macrocytosis and macro-ovalocytosis of red cells and hypersegmented neu-

Dow

nloa

ded

by:

149.

126.

78.4

9 -

11/1

9/20

15 1

0:11

:04

PM



trophils were predominant morphological changes in pe-ripheral blood. The bone marrow smears from the aspi-rates showed frank megaloblastosis of erythroid precur-sor cells, which were mainly characterized by larger cell size, and a fi nely stippled lacy pattern of nuclear chroma-tin, with relatively abundant cytoplasm and asynchrony between nuclear and cytoplasmic maturity ( fi g. 1 a) as compared to normoblasts (from bone marrows of control subjects) ( fi g. 1 b). The erythroid precursor cells in the bone marrow of patients with myelodysplastic syndrome and erythroleukemia also displayed variable degrees of immaturity of nuclear chromatin showing a fi nely stip-pled lacy pattern ( fi g. 1 c, d), resembling the megaloblasts in vitamin-B 12 - or folate-defi cient megaloblastic ane-mia.

Transmission Electron Microscopy These cytomorphological differences between the nor-

moblasts and megaloblasts appeared magnifi ed in the trans-mission electron microscopy with particular reference to the spongy, fi nely stippled appearance of the nuclear chro-matin in the megaloblasts in contrast to condensed chro-matin in the nuclei of normoblasts ( fi g. 2 a–d).

Cytogenetic Studies No signifi cant numerical changes were observed; the

model number of chromosomes was 46 and a very small number of metaphase plates showed random aneuploidy. However, the individual chromosomes showed striking structural changes, which were characterized by elonga-tion and despiralization of chromosomes ( fi g. 3 a, b);occasional polyploidy and endoreduplication were en-countered. The replication pattern of chromosomes in megaloblastic anemia also appeared different from nor-moblastic bone marrows as studied by autoradiography after incorporation of 3 H-TdR into DNA ( fi g. 4 a, b).

Cellular DNA Content, and DNA Synthesis (i.e., 3 H-TdR Incorporation into DNA) The measurement of DNA content of megaloblasts and

normoblasts by image analysis after Feulgen staining showed that the interphase (Go/G1) cells constituted the majority with modal DNA content of diploid cells ( fi g. 2 c). A relatively small proportion of the cell population ( ! 10%) showed random aneuploidy on either side of the diploid mode. Autoradiographic studies after 1 h of pulse expo-sure to 3 H-TdR in short-term culture of bone marrow cells showed that the labeling index as well as grain count of basophilic and polychromatic erythroblasts were signifi -cantly higher in the megaloblastic bone marrows than in the normal bone marrows ( table 3 ); the orthochromatic

Fig. 1. Light microscopic cytomorphological features of megalo-blasts ( a ) and normoblasts ( b ). Megaloblasts show relatively larger cell size, a fi nely stippled lacy pattern of nuclear chromatin, with asynchrony between nuclear and cytoplasmic maturation. In con-trast, normoblasts show condensed and clumpy nuclear chromatin. Bone marrows from patients with myelodysplastic syndrome ( c ) and erythroleukemia ( d ) also show megaloblastic changes.

Table 3. Results of autoradiographic studies for incorporation of 3H-TdR into DNA in bone marrow cells in vitro

Labelingindex

Meangrain count

Basophilic + polychromaticnormoblasts 2586.7 21.5785.3

Basophilic + polychromaticmegaloblasts 51.388.2* 43.688.7*

Control subjects: n = 30; vitamin-B12-defi cient subjects: n = 40; folate-defi cient subjects: n = 10. * p ̂ 0.01 compared to the normal values.

Dow

nloa

ded

by:

149.

126.

78.4

9 -

11/1

9/20

15 1

0:11

:04

PM


Fig. 2. Transmission electron microscopic picture of normoblasts ( a , b ) shows con-densed and clumpy chromatin; in contrast, transmission electron microscopic picture of megaloblasts ( c , d ) shows well-dispersed fi nely stippled rarefi ed nuclear chromatin.

Fig. 3. A metaphase plate from bone mar-row of a control (normal) subject ( a ) and from a patient with megaloblastic anemia ( b ) showing elongation and despiralization of individual chromosomes.

Fig. 4. Autoradiographic study of chromo-somal preparations after 3 H-TdR incorpo-ration showing the distribution pattern of silver grains due to 3 H-TdR uptake in nor-mal ( a ) and megaloblastic chromosomes ( b ).

Dow

nloa

ded

by:

149.

126.

78.4

9 -

11/1

9/20

15 1

0:11

:04

PM



erythroblasts did not take up 3 H-TdR as these cells do not proliferate. The cells of granulocytic series in megaloblas-tic anemia also showed relative slowing of maturation. The metamyelocytes in the control bone marrows did not incorporate 3 H-TdR, but giant metamyelocyte in the meg-aloblastic bone marrow showed incorporation of 3 H-TdR into DNA, indicating its proliferative ability ( fi g. 5 ).

dU Suppression Test The results of dU suppression tests in human bone

marrows are shown in table 4 . This has proved to be a very sensitive test to detect and quantitate megaloblasto-sis due to defi ciency of folate and vitamin B 12 , even in very mild or early clinical states when morphological manifestations remain equivocal [7, 10–12, 16, 17] . Fur-thermore, the dU suppression test was found to be normal when performed with bone marrows of patients with my-elodysplastic syndrome and erythroleukemia ( table 4 ) in which cytomorphological changes very similar to mega-loblastosis also occurred.

Morphology and DNA Synthesis in the Bone Marrows of Folate-Defi cient Animals Folate-Defi cient Rhesus Monkeys. Five adult monkeys

fed a folate-poor diet for 4–5 months developed hemato-logical and biochemical evidences of folate defi ciency. They developed mild to moderate anemia and thrombo-cytopenia with or without leukopenia. The peripheral blood smear showed macrocytic red cells and hyperseg-mented neutrophils. Serum and red cell folate as well as liver folate were markedly reduced as compared to the control monkeys, which were fed a folate-rich diet ( ta-ble 2 ). The serum vitamin B 12 levels remained normal.

The bone marrow smears of the control monkeys feda folate-rich diet showed normoblastic erythropoiesis ( fi g. 6 a), whereas the folate-defi cient monkeys revealed frank megaloblastic transformation ( fi g. 6 b). The dU sup-pression test of their aspirated bone marrows showed ab-normal results, which were corrected in vitro by the ad-dition of 5-methyl-THF but not by vitamin B 12 ( table 5 ). All of these features were similar to those of human meg-aloblastic anemia due to folate defi ciency.

Folate-Defi cient Rats. Weanling Wistar rats fed a fo-late-poor diet developed persistently low levels of folate in their serum, red cells and liver in 12–16 weeks ( table 2 ). They also developed mild to moderate anemia, thrombo-cytopenia and leukopenia, but red cells were still normo-cytic and normochromic. The bone marrow smears of control rats ( fi g. 6 c) as well as of folate-defi cient rats ( fi g. 6 d) showed normoblastic erythropoiesis (i.e., ab-

Fig. 5. Autoradiography of megaloblastic bone marrows showing 3 H-TdR-labeled gi-ant metamyelocytes indicating their prolif-erative ability.

Fig. 6. a A bone marrow smear from a control rhesus monkey show-ing normoblastic erythropoiesis. b A bone marrow smear from a folate-defi cient rhesus monkey showing fl orid megaloblastic trans-formation. c A bone marrow smear from a control rat showing normoblastic erythropoiesis. d A bone marrow smear from an es-tablished folate-defi cient rat showing normoblastic erythropoiesis. Absence of megaloblastic transformation.

Dow

nloa

ded

by:

149.

126.

78.4

9 -

11/1

9/20

15 1

0:11

:04

PM


sence of megaloblastic transformation). DNA synthesis as indicated by dU suppression test was abnormal or im-paired, which was corrected by the addition of methyl-THF, not by vitamin B 12 ( table 5 ).

Transmission Electron Microscopy of DNA Fibers from Human Patients with Megaloblastic Anemia and from Control Subjects DNA fi bers from normoblastic bone marrows (i.e.,

bone marrows of control subjects) after extraction with-out proteinase K appeared tightly knotted or beaded at regular intervals yielding a wavy pattern ( fi g. 7 a). When DNA was extracted from these cells with proteinase K, which removed nuclear proteins including histone, the DNA fi bers appeared straight, elongated with disappear-ance of the knots and the wavy pattern ( fi g. 7 b). On the other hand, DNA fi bers extracted from megaloblastic bone marrow cells without proteinase K showed very few or no beads and knots and consequently looked relative-

ly straightened and relaxed ( fi g. 7 c) and comparable to the DNA fi bers extracted from these cells using proteinase K ( fi g. 7 d), suggesting that megaloblasts had a markedly re-duced nuclear histone content.

Histone Biosynthesis in the Bone Marrows of Control Subjects and Patients The incorporation of the radioactive amino acid 3 H-

tryptophan is a marker of total histone biosynthesis, whereas the incorporation of 3 H-lysine occurs preferen-tially into lysine-rich histones (i.e., H 1 , H 2 A, and H 2 B histones), and that of 3 H-arginine into arginine-rich his-tones (i.e., H 3 and H 4 histones). The results of these stud-ies are shown in table 6 . There was a decrease in total histone synthesis in the folate- and vitamin-B 12 -defi cient megaloblastic bone marrows as compared to those of con-trols and iron defi ciency anemia. This defect appeared to have predominantly affected the synthesis of arginine-rich histones. Similar impairment of histone biosynthe-

Table 4. Summary of results of dU suppression tests in short-term cultures of bone marrow cells

dU alone dU + 5-me-thyl-THF

dU +CNC bl

dU + CNC bl +5-methyl-THF

dU +folic acid

Normal (n = 30) 7.582.2 6.881.8 7.182.0 8.582.3 9.081.4Vitamin B12 defi cient* (n = 40) 75.5815.0 59.6812.7 46.5815.2 9.283.2 8.682.4Folate defi cient* (n = 10) 63.8818.5 8.983.2 61.9821.0 7.581.8 9.082.1Myelodysplastic syndrome (n = 15) 7.281.8 9.382.4 6.982.1 7.782.5 8.282.0Erythroleukemia (n = 5) 9.582.6 6.881.9 7.582.3 8.282.2 7.481.6

dU suppression values expressed as percentage incorporation of 3H-TdR into DNA without dU (see sections on Methods). CNC bl = cyanocobalamin.

* p < 0.01 compared with normal values. The values for myelodysplastic syndrome and erythroleukemia were not signifi cant compared to the normal values. Results are means 8 SD.

Table 5. Results of dU suppression test in folate-defi cient rats and rhesus monkeys

dU alone dU +folic acid

dU +5-methyl-THF

dU +vitamin B12

Control rats (n = 10) 10.288.2 9.282.6 9.582.2 8.682.8Folate-defi cient rats (n = 10) 38.588.5* 9.883.0* 8.982.8* 40.5812.5*Control monkeys (n = 5) 9.182.2 10.182.5 8.681.8 7.983.3Folate-defi cient monkeys (n = 5) 52.6815.6* 9.883.5* 10.082.9* 49.5811.8*

dU suppression values are expressed as percentage of 3H-TdR incorporation into DNA (3H-TdR incorpora-tion in replicate cultures without dU is taken as 100%).

* p < 0.01; folate-defi cient rats compared with control rats and folate-defi cient monkeys compared with control monkeys.

Dow

nloa

ded

by:

149.

126.

78.4

9 -

11/1

9/20

15 1

0:11

:04

PM



sis, particularly of arginine-rich histones, was also ob-served in the bone marrow of patients with myelodysplas-tic syndrome and erythroleukemia.

Discussion

Several hematological conditions are known to be as-sociated with megaloblastic changes in the bone marrow. The most common among these are defi ciency of vitamin

B 12 and folate, resulting in megaloblastic anemia, which can be completely corrected or cured by treating them with the defi cient vitamins. Megaloblastic changes also occur in the bone marrow of patients with myelodysplas-tic syndrome and erythroleukemia, which are preneoplas-tic and neoplastic conditions, respectively; megaloblasto-sis in these two latter conditions are not reversed by ther-apy with vitamin B 12 and folate [1, 7, 9] . Irrespective of their causes, megaloblastosis in all of these conditions as stated above are cytomorphologically similar, character-

Table 6. The incorporation of 3H-tryptophan, 3H-lysine and 3H-arginine into nuclear histones

3H-tryptophan 3H-lysine 3H-arginine

Control subjects (n = 20) 21,50081,510 11,44881,200 10,17581,125Iron defi ciency anemia (n = 20) 22,05481,325 10,95581,175 10,11581,096Vitamin-B12-defi cient megaloblastic anemia (n = 30) 16,45581,024* 7,9958875* 2,7858615**Folate-defi cient megaloblastic anemia (n = 10) 15,14581,086* 8,1558920* 3,2258865**Erythroleukemia (n = 5) 17,5088512* 9,5268726* 5,8688497*Myelodysplastic syndrome (n = 15) 16,9738995* 8,9728698* 6,2878586*

The fi gures indicate the radioactivity (disintegrations per minute) incorporated. Results are means 8 SD. * p < 0.05; ** p < 0.01, compared to control subjects.

Fig. 7. Transmission electron microscopic picture of DNA fi bers from a control (nor-mal) bone marrow without proteinase K treatment showing knotted and spiral wavy pattern ( a ); DNA fi ber from the control (normal) bone marrow after extraction with proteinase K to remove nuclear proteins (e.g. histones) showing that DNA fi bers lost their knotted, beaded appearance and ap-peared relaxed and straight ( b ), from the bone marrow of a patient with megaloblas-tic anemia without proteinase K treatment during extraction ( c ), and with proteinase K treatment ( d ); both c and d showed loss of knots and wavy pattern resulting in de-spiralized and relaxed appearance of the fi -bers, suggesting absence or reduction of nu-clear histones in the DNA of megaloblasts.

Dow

nloa

ded

by:

149.

126.

78.4

9 -

11/1

9/20

15 1

0:11

:04

PM


ized by the open lacy pattern and spongy nuclear chro-matin in both light and transmission electron microscop-ic observations. The distinctive structural changes in the chromatin of megaloblasts in patients with vitamin B 12 and folate defi ciency were found to be associated with several structural abnormalities of individual chromo-somes such as elongation and despiralization (unwind-ing) of chromosomes. These may cause a defect in pack-aging chromosomes inside the cells, possibly impairing cell growth and cell functions, although no signifi cant ab-normalities of chromosome number were observed and the modal chromosome number remained 46 (diploid) with no consistent or signifi cant aneuploidy [19, 27–29] . It has been suggested that the cytomorphological and cy-togenetic changes in megaloblastic anemia may be an ex-pression of deranged cellular DNA synthesis [7–12] .

Cellular DNA comprises four deoxynucleotide tri-phosphate building blocks, namely deoxyadenosine tri-phosphate, deoxyguanosine triphosphate, deoxycytosine triphosphate and dTTP. In the double-stranded DNA he-lix, adenine pairs with thymine (A=T) and guanine with cytosine (G=A). The base composition of DNA extracted from megaloblastic and normoblastic bone marrows has

been shown to be similar and the ratios of A:T and G:C are close to unity [30] . The synthesis of DNA in replicat-ing cells is regulated by a number of reaction loops some of which are directly or indirectly dependent on the avail-ability of folate coenzymes and vitamin B 12 ( fi g. 8 ). The relevant and important reactions include: homocysteine to methionine interconversion, which requires methyl-cobalamin (CH 3 -B 12 ) as an essential cofactor; in this reac-tion, 5-methyl-THF, the predominant form of plasma fo-late which is metabolically inactive, is also converted into an active compound, THF. This latter compound is, in turn, converted into 5,10-methylene-THF, which is an essential coenzyme in converting dUMP to dTMP by the enzyme thymidylate synthase. The dU suppression test provides a sensitive quantitative assessment of megalo-blastic anemia and can also reveal the type of vitamin defi ciency causing this anemia. This test is capable of dis-tinguishing megaloblastosis associated with myelodys-plastic syndrome and erythroleukemia from that result-ing from defi ciency of vitamin B 12 and folate, since the dU suppression test is normal in the former two condi-tions, whereas it is abnormal in defi ciency of vitamin B 12 and folate. These observations also militate against the

Fig. 8. A cartoon showing metabolic inter-relations between vitamin B 12 and folate with special reference to the de novo and salvage pathways of thymidylate synthesis (dUMP ] dTMP).

Dow

nloa

ded

by:

149.

126.

78.4

9 -

11/1

9/20

15 1

0:11

:04

PM



hypothesis of megaloblastosis being an expression of de-ranged cellular DNA synthesis, regardless of the causes [1, 7–9] . Megaloblastic changes in erythroleukemia and myelodysplastic syndrome are possibly caused by mecha-nisms unrelated to the dU suppression abnormality [18] . The results of our studies on experimental folate defi -ciency in rhesus monkeys and Wistar strain rats are of considerable interest and relevant to the possible relation-ship between deranged DNA synthesis and the develop-ment of megaloblastosis. Both rhesus monkeys and rats, when made folate defi cient, developed an abnormality in DNA synthesis as indicated by an abnormal dU suppres-sion test, but only folate-defi cient monkeys developed frank megaloblastosis, whereas folate-defi cient rats re-mained unequivocally normoblastic. These observations support the concept that while DNA synthetic abnormal-ity as revealed by dU suppression test is a sensitive index of folate and/or vitamin B 12 defi ciency, this abnormality alone may not be the sole molecular event leading to meg-aloblastosis. Megaloblastic changes in the bone marrow also occur following chemotherapy with several drugs that impair DNA synthesis; but the precise molecular mechanisms underlying the development of megaloblas-tosis in those circumstances have not been adequately investigated.

The elongation and despiralization of chromosomes in replicating megaloblasts [19, 27–29] appear to refl ect their characteristic cytomorphological features in the in-terphase nuclear chromatin as observed in light and trans-mission electron microscopy. The results of our studies on electron microscopic confi guration of DNA fi bers ex-tracted from megaloblastic and normoblastic bone mar-rows appear to provide additional data on molecular events associated with megaloblastosis. The extraction of histones from DNA of normal bone marrow cells by treat-ment with proteinase K during extraction procedure made the DNA fi bers lose the knotted and spiral pattern, and appear straight with bundles of fi bers running paral-lel as observed by transmission electron microscopy; on the other hand, DNA fi bers extracted from megaloblastic bone marrows, even when untreated by proteinase K, ap-peared relatively straight and devoid of knots and without wavy or spiral pattern. These fi ndings suggest that the DNA extracted from megaloblastic bone marrows even without proteinase K treatment had a markedly decreased histone content as compared to DNA from normoblastic bone marrows ( fi g. 7 a–d). These assumptions were fur-ther supported by our studies on histone biosynthesis. The results of incorporation of 3 H-tryptophan, 3 H-lysine and 3 H-arginine into nuclear histone in short-term cul-

tures of megaloblastic as well as control (normoblastic) bone marrows provided a differential profi le ( table 6 ). Biosynthesis of histones in megaloblastic bone marrows was signifi cantly lower than in control bone marrows or in bone marrows of patients with iron defi ciency anemia. This reduction was more marked in respect of incorpora-tion of 3 H-arginine than in respect of 3 H-lysine and 3 H-tryptophan. Nuclear histones play a signifi cant role in coiling and spiralization of chromosomes, and consider-able attention has generally been focused on the role of these proteins in the regulation of the structure of chro-mosomes in replicating cells and chromatin in the inter-phase nucleus. There are several species of histones, some of which are lysine rich (H 1 , H 2 A, H 2 B) and others are arginine rich (H 3 , H 4 ). The nuclear chromatin is made up of repeating units (nucleosomes), each containing ap-proximately 200 bp of DNA and two each of H 2 A, H 2 B, H 3 and H 4 . Most of the DNA molecules are winded around the outside of a core of histones – the remaining DNA, the linker DNA, joins adjacent nucleosomes – and contribute to the fl exibility of the chromatin network. Thus, a chromatin is delicately joined to a chain of nu-cleosomes rather like beads on a string [21–23] . Changes in the composition of histones have been described in developing sperms [22] and duck erythroblasts [24] , and in both instances, arginine-rich histones have been shown to increase as a function of cell maturation. The decreased synthesis of predominantly arginine-rich histones in the megaloblasts may be related to their impaired nuclear maturation and asynchrony of nuclear cytoplasmic devel-opment. It may be suggested that cytomorphological changes of megaloblastosis may be closely related to an imbalance in the molecular species of histones, which in some instances (as in vitamin B 12 and folate defi ciency) may interact with slowly synthesizing DNA in the repli-cating cells. The reduced synthesis of particularly argi-nine-rich histone may be a common denominator of bio-chemical events in clinical and hematological conditions associated with megaloblastic transformation of erythro-poiesis. In recent years, several groups of investigators observed signifi cant associations of specifi c histone pat-terns, acetylase/deacetylase activities with cell cycle changes, and apoptosis in many neoplastic cell models [31, 32] . Intranuclear mechanisms that signal apoptosis after DNA damage overlap with those that initiate cell cycle arrest and disturb cell cycle checkpoints [32, 33] . Interactions between these nuclear events and apoptotic processes may cause genomic instability in other cell sys-tems as well [33, 34] . Naturally occurring or synthetic inhibitors of histone deacetylases can cause cell cycle ar-

Dow

nloa

ded

by:

149.

126.

78.4

9 -

11/1

9/20

15 1

0:11

:04

PM


rest, differentiation and apoptosis in neoplastic cells opening a door to explore newer cancer chemotherapy strategies [34, 35] . It is possible that changes in nuclear histones coupled with altered chromatin may also impair regulation of cell cycle checkpoints promoting apoptosis in megaloblastic bone marrows as suggested by several recent studies [36–39] . Recently, it has been reported that the nuclear histone content of bone marrows in lower mammalian species such as rats is much higher than that of higher mammalian species such as rhesus monkey [40] . This is believed to be a phylogenetic characteristic. When folate defi ciency was induced in monkeys, their bone marrow developed fl orid megaloblastic transformation, and this was associated with marked reduction of par-ticularly arginine-rich histones, but folate-defi cient rats did not develop megaloblastic transformation, although DNA synthesis was impaired as indicated by abnormal dU suppression values. This can be explained by the ini-tially high histone content in the bone marrow cells in the rats, which was not suffi ciently reduced even in estab-lished folate defi ciency to reach a critically low level re-quired for megaloblastic transformation [36] . A postu-lated scheme of molecular events in the development of megaloblastosis is given in fi gure 9 .

Conclusions

Cytomorphological features of megaloblasts in light and electron microscopy and the chromosomal pattern suggest that elongation, despiralization and uncoiling of chromosomes were related to the fi nely stippled lacy pat-

tern of the nuclear chromatin which distinguishes them from normoblasts. DNA synthetic abnormality as re-vealed by abnormal dU suppression test in vitamin-B 12 - and folate-defi cient megaloblastic anemia is not a consis-tent molecular event underlying the development of meg-aloblastosis in all clinical and hematological disorders in which megaloblastic changes are observed (e.g. myelodys-plastic syndrome and erythroleukemia). Nuclear pro-teins, particularly histones, are known to play a signifi -cant role in causing spiralization, twisting or coiling of chromosomes in the replicating cells and in the condensa-tion of nuclear chromatin in the interphase. The reduced biosynthesis of histones particularly of arginine-rich his-tones as shown in ‘megaloblastic’ bone marrows irrespec-tive of their causes appears to be a common denominator of molecular events associated with the development of megaloblastosis.

Acknowledgements

This work was supported by Kuwait University Research Grant No. MG 011 and the Institute of Immunohematology, Mumbai, India. The authors highly appreciate the excellent work done and the painstaking care taken by Mr. James Luke, Secretary, Depart-ment of Pathology, in preparing the manuscript.

Fig. 9. A cartoon showing the postulated molecular basis of development of megalo-blastosis.

Dow

nloa

ded

by:

149.

126.

78.4

9 -

11/1

9/20

15 1

0:11

:04

PM



References

1 Chanarin I: The Megaloblastic Anaemia, ed 3. Oxford, Blackwell Scientifi c Publications, 1990, pp 315–331.

2 Wills L: Treatment of pernicious anaemia of pregnancy and ‘tropical anaemia’ with special reference to yeast extract as a curative agent. Br Med J 1931; 1: 1059–1064.

3 Wills L: Treatment of pernicious anaemia, nu-tritional macrocytic anaemia and tropical sprue. Blood 1948; 3: 36–56.

4 Chatterjea JB: Nutritional megaloblastic anae-mia in tropical zones. Proc 11th Congr Int Soc Hematol, Sidney, 1966, pp 120–132.

5 Das KC: Nutritional megaloblastic anaemia. Proc Symp Nutr Anemias, Plenary Sess 2, Meet Asian Pacifi c Div, Int Soc Hematol, Mel-bourne, 1971, pp 11–13.

6 Herbert V: The fi ve possible causes of all nutri-ent defi ciency. Aust NZ J Med 1972; 1: 69–72.

7 Das KC, Herbert V: Vitamin B 12 -folate inter-relations. Clin Haematol 1976; 5: 997–725.

8 Herbert V: The Megaloblastic Anemias, ed 1. New York, Grune and Stratton, 1959, p 1.

9 Lee RG: Megaloblastic anemias: Disorders of impaired DNA synthesis; in Lee RG, Foester J, Lukens J, Parakevas JP, Rodgers GM (eds): Wintrobe’s Clinical Hematology, ed 10. Balti-more, Williams & Wilkins, 1999, pp 941–978.

10 Das KC, Hoffbrand AV: Lymphocyte transfor-mation in megaloblastic anaemias. Morphol-ogy and DNA synthesis. Br J Haematol 1970;

19: 459–468. 11 Das KC, Herbert V: The lymphocyte as a

marker of past nutritional status. Persistence of abnormal lymphocyte deoxyuridine (dU) suppression test and chromosomes in patients with past defi ciency of folate and vitamin B 12. Br J Haematol 1978; 38: 219–223.

12 Metz J, Kelly A, Swett VC, Waxman S, Herbert V: Deranged DNA synthesis by bone marrow from vitamin B 12 -defi cient humans. Br J Hae-matol 1968; 14: 575–592.

13 Mohanty D, Das KC: Effect of folate defi cien-cy on the reproductive organs of female Rhesus monkeys: A cytomorphological and cytokinet-ic study. J Nutr 1982; 112: 1505–1576.

14 Johnson EM, Nelson MM, Monie IN: The composition of folate-defi cient diet for rats. Anat Rec 1963; 146: 215.

15 Dacie JV, Lewis SM: Practical Haematology, ed 8. London, Churchill Livingstone, 1995, pp 49–143.

16 Das KC, Herbert V: In vitro DNA synthesis by megaloblastic bone marrow: Effect of folates and cobalamins on thymidine incorporation and de novo thymidylate synthesis. Am J He-matol 1989; 31: 11–20.

17 Das KC, Manusselis C, Herbert V: In vitro DNA synthesis by bone marrow cells and PHA-stimulated lymphocytes. Suppression by non radioactive thymidine of the incorpora-tion of 3 H-deoxyuridine into DNA: Enhance-ment of incorporation when inadequate vita-min B 12 and folate is corrected. Br J Haematol 1980; 44: 51–63.

18 Das KC, Garewal G, Mohanty D: Derange-ment of DNA synthesis in erythroleukemia. Normal deoxyuridine suppression and im-paired thymidine incorporation in bone mar-row culture. Acta Haematol 1980; 64: 121–134.

19 Das KC, Mohanty D, Garewal G: Cytogenetics in nutritional megaloblastic anaemia: Pro-longed persistence of chromosomal abnormal-ities in lymphocytes after remission. Acta Hae-matol 1986; 76: 146–154.

20 Sambrook J, Fritsch EF, Maniatis J: Molecular Cloning: A Laboratory Manual. New York, Cold Spring Harbor Laboratory Press, 1989, pp 1–25.

21 Murray K: The acid extraction of histones from calf thymus deoxyribonucleoprotein. J Mol Biol 1966; 15: 409–419.

22 Das KC, Kaufman RP, Gay H: Autoradio-graphic evidence of synthesis of an arginine-rich histone during spermatogenesis in Dro-sophila melanogaster. Nature 1964; 204:

1008–1009. 23 Tidwell T, Allfrey VF, Mirsky AE: The meth-

ylation of histone during regeneration of the liver. J Biol Chem 1968; 243: 707–715.

24 Gershey FL, Halett GW, Vidali G, Allfrey VG: Chemical studies on histone methylation. Evi-dence for the occurrence of 3-methyl-histidine in avian erythrocyte histone fraction. J Biol Chem 1969; 244: 4871–4877.

25 Longo DL, Herbert V: Radioassay for serum and red cell folate. J Lab Clin Med 1976; 87:

138–151. 26 Das KC, Manusselis, Herbert V: Determina-

tion of vitamin B 12 (cobalamin) in serum and erythrocytes by radioassay and of holo-trans-cobalamin II ( holo-TC-II) and holo-haptocor-rin (holo-TC1 and III) in serum by adsorbing holo-TC-II on microfi ne silica. J Nutr Biochem 1991; 2: 455–463.

27 Kiossoglou, KA, Mitus WJ, Dameshek W: Chromosomal aberrations in pernicious ane-mia. Study of three cases before and after ther-apy. Blood 1965; 5: 662–682.

28 Heath CW: Cytogenetic observations in vita-min B 12 and folate defi ciency. Blood 1966; 27:

800–815. 29 Das KC: Nutritional megaloblastosis. Symp

Nutr Anaemias. Proc Second Congr Asian Pa-cifi c Div, Int Soc Haematol, New Delhi, 1976, pp 14–15.

30 Hoffbrand AV, Pegg AE: Base composition of normal and megaloblastic bone marrow DNA. Nat New Biol 1972; 235: 187–188.

31 Khobta A, Carlo-Stella C, Capranico G: Spe-cifi c histone patterns and acetylase/deacetylase activity at the breakpoint-cluster region of the human MLL gene. Cancer Res 2004; 64: 2656–2662.

32 Zhao J: Coordination of DNA synthesis and histone gene expression during normal cell cy-cle progression and after DNA damage. Cell Cycle 2004; 3: 695–697.

33 Norbury CJ, Zhivolovsky B: DNA damage-induced apoptosis. Oncogene 2004; 23: 2797–2808.

34 Wang C, Fu M, Pestell RG: Histone acetyla-tion/deacetylation as a regulator of cell cycle gene expression. Methods Mol Biol 2004; 241:

207–216. 35 McLaughlin F, La Thangue NB: Histone

deacetylase inhibitors open new doors in can-cer therapy. Biochem Pharmacol 2004; 68:

1139–1144. 36 Ingram CF, Davidoff AN, Marais F, Sherman

GG, Mendelow BV: Evaluation of DNA analy-sis for evidence of apoptosis in megaloblastic anaemia. Br J Haematol 1997; 96: 576–583.

37 Koury MJ, Price JO, Hicks GG: Apoptosis in megaloblastic anemia occurs during DNA syn-thesis p53-independent, nucleoside-reversible mechanism. Blood 2000; 96: 3249–3255.

38 Gu L, Wu J, Qiu L, Jennings CD, Li GM: In-volvement of DNA mismatch repair infolate defi ciency-induced apoptosis small star, fi lled. J Nutr Biochem 2002; 13: 355–363.

39 Li GM, Presnell SR, Gu I: Folate defi ciency, mismatch repair-independent apoptosis and human disease. J Nutr Biochem 2003; 14: 568–575.

40 Das KC, Easow SK, Jadaon M, Das M, Mo-hanty D: The development of megaloblastosis is a species-restricted phenomenon. 9th Annu Health Sci Poster Conf, Kuwait, April 19–21, 2004, p 284.

D

ownl

oade

d by

: 14

9.12

6.78

.49

- 11

/19/

2015

10:

11:0

4 P

M

megaloblastosis: from morphos to...

Documents